Project Summary/Abstract Interneurons are a ubiquitous part of nearly every brain region that has been studied. They play a diverse and important set of functions, yet there is a gap between our understanding of these functions and their implementation at the level of cellular and circuit mechanisms. The antennal lobe is the first brain region in the Drosophila olfactory system, and has proven to be a valuable circuit for understanding the gap between neural computations and their implementation on a cellular and circuit level. The overarching goal of this proposal is to investigate how circuit mechanisms give rise to specific functions of inhibition ? such as gain control and temporal filtering. The specific hypothesis of this proposal is that antennal lobe local neurons (LNs) with different physiological properties connect to distinct synaptic targets and perform qualitatively different network functions. Alternatively, different LNs may have essentially the same network function, just in slightly different regimes (e.g., acting on fast versus slow timescales), or else functions of inhibition may be emergent properties of the entire LN cohort. The proposed approach will consider all three alternatives, any of which would be of interest to the field. Aim 1 will establish correlations between LN odor response physiology (e.g., ON/OFF, fast/slow, concentration variant/invariant) and LN synaptic connectivity. In vivo whole-cell recordings will be performed to determine which physiological LN types map onto which morphological LN types. The use of existing serial section electron microscopy data (ssEM) will then aid in mapping physiological LN types directly onto synaptic connectivity profiles, by using morphological LN types as the link. The synthesis of these two data sets will determine how different physiological LN types connect to different targets. Aim 2 will investigate how the antennal lobe network is affected by activating/silencing different LN types. In Aim 2a in vivo current-clamp recordings will be performed from antennal lobe output projection neurons (PNs) and LNs while presenting odor stimuli, both with and without optogenetic activation/inactivation of specific LN types. The following metrics will be used to determine how boosting/suppressing the activity of specific LN types alter: (1) the ability of PNs to maintain concentration invariant odor representations (gain control), (2) the ability of PNs to transmit odor information over a wide range of timescales (temporal filtering), and (3) the properties of odor- evoked oscillations. In Aim 2b a simple firing rate model of the antennal lobe will be constructed with connection weights taken from existing ssEM reconstructions. Similar activation/inactivation experiments will then be performed on model LNs, and PN performance will be assessed using the same three metrics described above. The combination of these methods uniquely positions this project to develop a clear conceptual framework for understanding how specific interneuron types interact in a network. The behavior of interneuron networks is a topical problem with broad relevance, and this study will have significance for a wide community of neurobiologists. Moreover, understanding how interneurons function in a network will be critical for determining how to treat psychiatric disorders that are linked to interneuron dysfunction, such as schizophrenia, autism spectrum disorders, and epilepsy.